Bug#664589: xfonts-encodings: Please mark as Multi-Arch: foreign
Package: xfonts-encodings
Version: 1:1.0.4-1
Severity: minor
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu ubuntu-patch precise
Hello,
In https://launchpad.net/bugs/957045 it was reported that
xfonts-encodings causes uninstallable build dependencies for some
multi-arch cross builds.
As per the Multi-Arch spec this package should be marked as foreign.
Simple debdiff attached, by Launchpad user wookey.
Thanks for considering,
Martin
--
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)
diff -u xfonts-encodings-1.0.4/debian/control
xfonts-encodings-1.0.4/debian/control
--- xfonts-encodings-1.0.4/debian/control
+++ xfonts-encodings-1.0.4/debian/control
@@ -20,4 +21,5 @@
Depends: ${shlibs:Depends}, ${misc:Depends}, x11-common
Replaces: xfonts-base ( 1:1.0.0)
+Multi-Arch: foreign
Description: Encodings for X.Org fonts
This package contains the encodings that map to specific characters.
diff -u xfonts-encodings-1.0.4/debian/changelog
xfonts-encodings-1.0.4/debian/changelog
--- xfonts-encodings-1.0.4/debian/changelog
+++ xfonts-encodings-1.0.4/debian/changelog
@@ -1,3 +1,10 @@
+xfonts-encodings (1:1.0.4-1ubuntu1) precise; urgency=low
+
+ * debian/control: Mark as Multi-Arch: foreign to allow using this as
+build dependency for cross-building. (LP: #957045)
+
+ -- Wookey woo...@wookware.org Thu, 15 Mar 2012 14:15:14 +
+
xfonts-encodings (1:1.0.4-1) unstable; urgency=low
[ Julien Cristau ]
signature.asc
Description: Digital signature
Bug#570447: Acknowledgement (x11-common: Optimize speed of Xsession.d scripts)
Hello Julien, Julien Cristau [2010-04-03 19:16 +0200]: we went from [...] to [...] The changelog says this was to avoid testing for xrdb if ~/.Xresources doesn't exist. But we'll keep testing for xrdb anyway, since we'll pretty much always have /etc/X11/Xresources/. So you'll still run the exact same tests. Ah, of course. I'm not sure now, but I guess back then I simply overlooked the system Xresource file then. So please feel free to revert that. Thanks, Martin -- Martin Pitt| http://www.piware.de Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org) signature.asc Description: Digital signature
Bug#570447: Acknowledgement (x11-common: Optimize speed of Xsession.d scripts)
Hello Julien, Julien Cristau [2010-03-14 12:04 +0100]: +- Use shell built in type instead of external which to test for + programs. There's no guarantee that /bin/sh has a 'type' built-in, as far as I can tell from SUSv3 and policy). I'd suggest command -v, but apparently posh doesn't like that either, so I don't know. Hm, it's present in dash and bash, anyway. But if you like command -V better, I'm fine with that. +- 30x11-common_xresources: Swap the order of tests to keep the most + unlikely (like ~/.Xresources exists) outside, to avoid running the + other tests (like xrdb exists) on systems which don't use Xresources. + I'm not sure about this one. x11-common installs a file in /etc/X11/Xresources, so you'll end up looking for xrdb on pretty much every system regardless. Right, it's a minor case, if someone decides to remove /etc/X11/Xresources/x11-common. But I still think it's a tad less likely to be true than the existence of xrdb. And then the reordering means you're looking for it twice (granted, with 'type' the shell can cache the result of the first lookup, but still). Hm, I don't understand? The second type is only done if [ -f $USRRESOURCES ], and ~/.Xresources should exist pretty seldomly these days? Thanks, Martin -- Martin Pitt| http://www.piware.de Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org) signature.asc Description: Digital signature
Bug#572692: Please provide udev rule instead of hal fdi
Package: xserver-xorg-input-vmmouse
Version: 1:12.6.5-2
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu lucid ubuntu-patch
Hello!
vmmouse currently does not work out of the box any more, since X.org
does not use hal any more for detecting input devices.
I created an udev rule and tested it in both kvm (with vmmouse driver)
as well as real iron (confirmed that it doesn't apply there), works
well.
debdiff attached, which also cares about removing the hal bits and
triggering udev.
I also sent the rule upstream:
https://bugs.freedesktop.org/show_bug.cgi?id=26910
Thanks,
Martin
--
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)
diff -u
xserver-xorg-input-vmmouse-12.6.5/debian/xserver-xorg-input-vmmouse.install
xserver-xorg-input-vmmouse-12.6.5/debian/xserver-xorg-input-vmmouse.install
--- xserver-xorg-input-vmmouse-12.6.5/debian/xserver-xorg-input-vmmouse.install
+++ xserver-xorg-input-vmmouse-12.6.5/debian/xserver-xorg-input-vmmouse.install
@@ -3,3 +3,2 @@
-usr/lib/hal/hal-probe-vmmouse
usr/share/man
-usr/share/hal/fdi/policy/20thirdparty/*
+../69-xorg-vmmouse.rules /lib/udev/rules.d
diff -u xserver-xorg-input-vmmouse-12.6.5/debian/changelog
xserver-xorg-input-vmmouse-12.6.5/debian/changelog
--- xserver-xorg-input-vmmouse-12.6.5/debian/changelog
+++ xserver-xorg-input-vmmouse-12.6.5/debian/changelog
@@ -1,3 +1,14 @@
+xserver-xorg-input-vmmouse (1:12.6.5-2ubuntu4) lucid; urgency=low
+
+ * Add debian/69-xorg-vmmouse.rules to automatically use vmmouse on
+appropriate devices, for udevified X servers. (LP: #430532)
+ * debian/xserver-xorg-input-vmmouse.install: Do not install the hal fdi and
+callout any more; install the udev rule.
+ * debian/xserver-xorg-input-vmmouse.postinst.in: Trigger a change event on
+mouse input devices, to get the new/updated rules applied.
+
+ -- Martin Pitt martin.p...@ubuntu.com Fri, 05 Mar 2010 18:29:00 +0100
+
xserver-xorg-input-vmmouse (1:12.6.5-2ubuntu3) lucid; urgency=low
* debian/patches/enable-detect-in-kvm.patch: add iopl() back so
only in patch2:
unchanged:
---
xserver-xorg-input-vmmouse-12.6.5.orig/debian/xserver-xorg-input-vmmouse.postinst.in
+++
xserver-xorg-input-vmmouse-12.6.5/debian/xserver-xorg-input-vmmouse.postinst.in
@@ -0,0 +1,17 @@
+#!/bin/sh
+set -e
+
+THIS_PACKAGE=xserver-xorg-input-evdev
+THIS_SCRIPT=postinst
+
+#INCLUDE_SHELL_LIB#
+
+
+if [ $1 = configure ]; then
+# update udev for our new/changed rules
+udevadm trigger --property-match=ID_INPUT_MOUSE=1 --action=change || true
+fi
+
+#DEBHELPER#
+
+exit 0
only in patch2:
unchanged:
--- xserver-xorg-input-vmmouse-12.6.5.orig/debian/69-xorg-vmmouse.rules
+++ xserver-xorg-input-vmmouse-12.6.5/debian/69-xorg-vmmouse.rules
@@ -0,0 +1 @@
+ACTION==add|change, ENV{ID_INPUT_MOUSE}==?*, ATTRS{description}==i8042
AUX port, PROGRAM=/usr/bin/vmmouse_detect, ENV{x11_driver}=vmmouse
signature.asc
Description: Digital signature
Bug#572692: Please provide udev rule instead of hal fdi
Hello Julien,
Julien Cristau [2010-03-05 20:13 +0100]:
Thanks for the patch, this should be fine for sid/squeeze.
Thanks. This should probably fix the package name typo in postinst.in
(although it should be harmless).
For upstream, the libudev backend of the server only looks for devices
with ID_INPUT set
That part of the rule should be fine, though. ID_INPUT_MOUSE is set by
udev itself, and it's just for discovering which device to set the
driver for.
, the x11_driver property is ignored (the driver is set from
xorg.conf / xorg.conf.d).
Right, that would be a problem indeed. I keep forgetting about this.
Currently the configuration can match devices by name, vendor,
device path, the ID_INPUT_foo attributes, and an ID_INPUT.tags
attribute. I'm not sure which would be more appropriate here, maybe
setting a vmware tag on some device...
Right, the rule might do ENV{ID_INPUT.vmmouse}=1 instead of setting
x11_driver?
Thanks,
Martin
--
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)
signature.asc
Description: Digital signature
Bug#570447: Acknowledgement (x11-common: Optimize speed of Xsession.d scripts)
Brice Goglin [2010-02-19 12:37 +0100]: How faster is the boot thanks to this patch ? It's nothing serious, of course, in the 0.1 s range. I was just going through all Xsession.d scripts, and some like 90consolekit were quite heavy. So I just cleaned up all of them while I was at it. Now the entire Xsession.d collection just has one external program call (the cat of the option file at the very beginning, thanks to dash not knowing about the $( file) syntax that bash supports). Martin -- Martin Pitt| http://www.piware.de Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org) -- To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100219125429.ga2...@piware.de
Bug#570447: x11-common: Optimize speed of Xsession.d scripts
Package: xorg
Version: 7.5+3
Severity: wishlist
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu lucid ubuntu-patch
Hello,
In our vendetta for faster boot I examined x11-common's Xsession.d
scripts and noticed that they call a lot of external programs (grep
several times, which) and run some code which is unlikely to be useful
on many machines (such as ~/.Xresources).
Attached patch streamlines the scripts to prefer shell builtins over
external programs. (Details see changelog).
Thanks for considering!
Martin
--
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)
diff -Nru xorg-7.5+1ubuntu2/debian/changelog xorg-7.5+1ubuntu3/debian/changelog
--- xorg-7.5+1ubuntu2/debian/changelog 2010-01-26 00:38:13.0 +0100
+++ xorg-7.5+1ubuntu3/debian/changelog 2010-02-18 23:14:11.0 +0100
@@ -1,3 +1,18 @@
+xorg (1:7.5+1ubuntu3) lucid; urgency=low
+
+ * Improve startup speed of Xsession.d scripts by eliminating all unnecessary
+external program calls:
+- In 20x11-common_process-args, cat $OPTIONFILE once into a variable and
+ use POSIX variable substitution in all scripts instead of calling grep
+ for every single test.
+- Use shell built in type instead of external which to test for
+ programs.
+- 30x11-common_xresources: Swap the order of tests to keep the most
+ unlikely (like ~/.Xresources exists) outside, to avoid running the
+ other tests (like xrdb exists) on systems which don't use Xresources.
+
+ -- Martin Pitt martin.p...@ubuntu.com Thu, 18 Feb 2010 23:02:22 +0100
+
xorg (1:7.5+1ubuntu2) lucid; urgency=low
[ David Planella ]
diff -Nru xorg-7.5+1ubuntu2/debian/local/Xsession.d/20x11-common_process-args
xorg-7.5+1ubuntu3/debian/local/Xsession.d/20x11-common_process-args
--- xorg-7.5+1ubuntu2/debian/local/Xsession.d/20x11-common_process-args
2010-01-25 10:53:09.0 +0100
+++ xorg-7.5+1ubuntu3/debian/local/Xsession.d/20x11-common_process-args
2010-02-18 23:13:59.0 +0100
@@ -2,6 +2,10 @@
# This file is sourced by Xsession(5), not executed.
+# read OPTIONFILE
+OPTIONS=$(cat $OPTIONFILE) || true
+
+
# Determine how many arguments were provided.
case $# in
0)
@@ -12,7 +16,7 @@
case $1 in
failsafe)
# Failsafe session was requested.
-if grep -qs ^allow-failsafe $OPTIONFILE; then
+if [ ${OPTIONS#*allow-failsafe} != $OPTIONS ]; then
if [ -e /usr/bin/x-terminal-emulator ]; then
if [ -x /usr/bin/x-terminal-emulator ]; then
exec x-terminal-emulator -geometry +1+1
diff -Nru xorg-7.5+1ubuntu2/debian/local/Xsession.d/30x11-common_xresources
xorg-7.5+1ubuntu3/debian/local/Xsession.d/30x11-common_xresources
--- xorg-7.5+1ubuntu2/debian/local/Xsession.d/30x11-common_xresources
2008-06-25 04:05:17.0 +0200
+++ xorg-7.5+1ubuntu3/debian/local/Xsession.d/30x11-common_xresources
2010-02-18 22:58:38.0 +0100
@@ -4,25 +4,23 @@
# If xrdb (from xbase-clients) is installed, merge system-wide X resources.
# Then merge the user's X resources, if the options file is so configured.
-if /usr/bin/which xrdb /dev/null 21; then
- if [ -d $SYSRESOURCES ]; then
-RESOURCEFILES=$(run-parts --list $SYSRESOURCES)
-if [ -n $RESOURCEFILES ]; then
- for RESOURCEFILE in $RESOURCEFILES; do
-xrdb -merge $RESOURCEFILE
- done
-fi
+if [ -d $SYSRESOURCES ] type xrdb /dev/null 21; then
+ RESOURCEFILES=$(run-parts --list $SYSRESOURCES)
+ if [ -n $RESOURCEFILES ]; then
+for RESOURCEFILE in $RESOURCEFILES; do
+ xrdb -merge $RESOURCEFILE
+done
fi
+fi
- if grep -qs ^allow-user-resources $OPTIONFILE; then
-if [ -f $USRRESOURCES ]; then
- xrdb -merge $USRRESOURCES
-fi
+if [ ${OPTIONS#*allow-user-resources} != $OPTIONS ] [ -f
$USRRESOURCES ]; then
+ if type xrdb /dev/null 21; then
+xrdb -merge $USRRESOURCES
+ else
+# Comment out this command if you desire a legacy-free X environment, and
find
+# the warning spurious.
+message warning: xrdb command not found; X resources not merged.
fi
-else
- # Comment out this command if you desire a legacy-free X environment, and
find
- # the warning spurious.
- message warning: xrdb command not found; X resources not merged.
fi
# vim:set ai et sts=2 sw=2 tw=80:
diff -Nru
xorg-7.5+1ubuntu2/debian/local/Xsession.d/50x11-common_determine-startup
xorg-7.5+1ubuntu3/debian/local/Xsession.d/50x11-common_determine-startup
--- xorg-7.5+1ubuntu2/debian/local/Xsession.d/50x11-common_determine-startup
2008-06-25 04:05:17.0 +0200
+++ xorg-7.5+1ubuntu3/debian/local/Xsession.d/50x11-common_determine-startup
2010-02-18 22:59:32.0 +0100
@@ -7,7 +7,7 @@
# executable, fall back to looking for a user's custom X session script, if
# allowed by the options file.
if [ -z $STARTUP ]; then
- if grep -qs ^allow-user-xsession $OPTIONFILE
Bug#570447: Acknowledgement (x11-common: Optimize speed of Xsession.d scripts)
Hello again,
argh, I just realized that the previous patch would catch commented
options as well, sorry for the blunder!
This updated patch defines a new function has_options which
encapsulates the option parsing, and thus makes the code easier to
read.
Thanks,
Martin
--
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)
diff -Nru xorg-7.5+1ubuntu2/debian/changelog xorg-7.5+1ubuntu3/debian/changelog
--- xorg-7.5+1ubuntu2/debian/changelog 2010-01-26 00:38:13.0 +0100
+++ xorg-7.5+1ubuntu3/debian/changelog 2010-02-18 23:14:11.0 +0100
@@ -1,3 +1,18 @@
+xorg (1:7.5+1ubuntu3) lucid; urgency=low
+
+ * Improve startup speed of Xsession.d scripts by eliminating all unnecessary
+external program calls:
+- In 20x11-common_process-args, cat $OPTIONFILE once into a variable and
+ use POSIX variable substitution in all scripts instead of calling grep
+ for every single test.
+- Use shell built in type instead of external which to test for
+ programs.
+- 30x11-common_xresources: Swap the order of tests to keep the most
+ unlikely (like ~/.Xresources exists) outside, to avoid running the
+ other tests (like xrdb exists) on systems which don't use Xresources.
+
+ -- Martin Pitt martin.p...@ubuntu.com Thu, 18 Feb 2010 23:02:22 +0100
diff -Nru xorg-7.5+1ubuntu2/debian/changelog xorg-7.5+1ubuntu4/debian/changelog
--- xorg-7.5+1ubuntu2/debian/local/Xsession.d/20x11-common_process-args
2010-01-25 10:53:09.0 +0100
+++ xorg-7.5+1ubuntu4/debian/local/Xsession.d/20x11-common_process-args
2010-02-19 06:27:51.0 +0100
@@ -2,6 +2,18 @@
# This file is sourced by Xsession(5), not executed.
+# read OPTIONFILE
+OPTIONS=$(cat $OPTIONFILE) || true
+
+has_option() {
+ if [ ${OPTIONS#*
+$1} != $OPTIONS ]; then
+return 0
+ else
+return 1
+ fi
+}
+
# Determine how many arguments were provided.
case $# in
0)
@@ -12,7 +24,7 @@
case $1 in
failsafe)
# Failsafe session was requested.
-if grep -qs ^allow-failsafe $OPTIONFILE; then
+if has_option allow-failsafe; then
if [ -e /usr/bin/x-terminal-emulator ]; then
if [ -x /usr/bin/x-terminal-emulator ]; then
exec x-terminal-emulator -geometry +1+1
diff -Nru xorg-7.5+1ubuntu2/debian/local/Xsession.d/30x11-common_xresources
xorg-7.5+1ubuntu4/debian/local/Xsession.d/30x11-common_xresources
--- xorg-7.5+1ubuntu2/debian/local/Xsession.d/30x11-common_xresources
2008-06-25 04:05:17.0 +0200
+++ xorg-7.5+1ubuntu4/debian/local/Xsession.d/30x11-common_xresources
2010-02-19 06:28:19.0 +0100
@@ -4,25 +4,23 @@
# If xrdb (from xbase-clients) is installed, merge system-wide X resources.
# Then merge the user's X resources, if the options file is so configured.
-if /usr/bin/which xrdb /dev/null 21; then
- if [ -d $SYSRESOURCES ]; then
-RESOURCEFILES=$(run-parts --list $SYSRESOURCES)
-if [ -n $RESOURCEFILES ]; then
- for RESOURCEFILE in $RESOURCEFILES; do
-xrdb -merge $RESOURCEFILE
- done
-fi
+if [ -d $SYSRESOURCES ] type xrdb /dev/null 21; then
+ RESOURCEFILES=$(run-parts --list $SYSRESOURCES)
+ if [ -n $RESOURCEFILES ]; then
+for RESOURCEFILE in $RESOURCEFILES; do
+ xrdb -merge $RESOURCEFILE
+done
fi
+fi
- if grep -qs ^allow-user-resources $OPTIONFILE; then
-if [ -f $USRRESOURCES ]; then
- xrdb -merge $USRRESOURCES
-fi
+if has_option allow-user-resources [ -f $USRRESOURCES ]; then
+ if type xrdb /dev/null 21; then
+xrdb -merge $USRRESOURCES
+ else
+# Comment out this command if you desire a legacy-free X environment, and
find
+# the warning spurious.
+message warning: xrdb command not found; X resources not merged.
fi
-else
- # Comment out this command if you desire a legacy-free X environment, and
find
- # the warning spurious.
- message warning: xrdb command not found; X resources not merged.
fi
# vim:set ai et sts=2 sw=2 tw=80:
diff -Nru
xorg-7.5+1ubuntu2/debian/local/Xsession.d/50x11-common_determine-startup
xorg-7.5+1ubuntu4/debian/local/Xsession.d/50x11-common_determine-startup
--- xorg-7.5+1ubuntu2/debian/local/Xsession.d/50x11-common_determine-startup
2008-06-25 04:05:17.0 +0200
+++ xorg-7.5+1ubuntu4/debian/local/Xsession.d/50x11-common_determine-startup
2010-02-19 06:28:57.0 +0100
@@ -7,7 +7,7 @@
# executable, fall back to looking for a user's custom X session script, if
# allowed by the options file.
if [ -z $STARTUP ]; then
- if grep -qs ^allow-user-xsession $OPTIONFILE; then
+ if has_option allow-user-session; then
for STARTUPFILE in $USERXSESSION $ALTUSERXSESSION; do
if [ -e $STARTUPFILE ]; then
if [ -x $STARTUPFILE ]; then
@@ -36,7 +36,7 @@
# If we still have not found a startup program, give up.
if [ -z $STARTUP ]; then
ERRMSG=unable to start X session ---
- if grep -qs
Bug#564258: xserver-xorg-input-evdev: evdev on accelerometers is broken
Hello Julien,
sorry for the late answer, was on a conference last week.
Julien Cristau [2010-01-08 21:37 +]:
On Fri, Jan 8, 2010 at 18:57:28 +, Julien Cristau wrote:
Apparently loading evdev on accelerometers results in a mouse cursor
stuck in the center of the screen. Not sure if we should somehow
blacklist these devices from the udev rules, or if the driver should be
smarter.
Peter, would you have any idea how to handle the above issue?
Martin, the kernel seems to assign a js (joystick?) handler to that
device. Should udev set ID_INPUT_JOYSTICK here?
I don't think it should. At least hal and udev define a joystick as an
input device which as absolute X/Y axes _and_ a button (BTN_TRIGGER,
BTN_A, or BTN_1). Supposedly the kernel is too sloppy here and just
checks for absolute axes?
E: SUBSYSTEM=input
E: ID_INPUT=1
E: x11_driver=evdev
I wonder which rule assigns this. At least in Ubuntu,
/lib/udev/rules.d/65-xorg-evdev.rules looks like
-- 8 --
ACTION!=add|change, GOTO=xorg_evdev_end
# By default, we use the -evdev driver for every mouse, keyboard, touchscreen
# or touchpad; later rules can then change the driver for specific input
# devices.
ENV{ID_INPUT}==, GOTO=xorg_evdev_end
ENV{ID_INPUT_MOUSE}==?*, ENV{x11_driver}=evdev
ENV{ID_INPUT_KEY}==?*, ENV{x11_driver}=evdev
ENV{ID_INPUT_TOUCHSCREEN}==?*, ENV{x11_driver}=evdev
ENV{ID_INPUT_TOUCHPAD}==?*, ENV{x11_driver}=evdev
LABEL=xorg_evdev_end
-- 8 --
I. e. it does _not_ assign evdev to any ID_INPUT device, just to
known types. In particular, it doesn't use evdev for joysticks either,
since the perception was that this is too confusing. Such a rule could
be shipped by xserver-xorg-input-joystick.
Do you happen to have a rule which looks like
ENV{ID_INPUT}==?*, ENV{x11_driver}=evdev
? If so, my recommendation would be to drop this. We don't need X.org
to listen to obscure devices like your accelerometer or headphone
jacks. Button-type devices like lids are already covered by
ID_INPUT_KEY (which is not the same as _KEYBOARD, NB).
Thanks,
Martin
--
Martin Pitt| http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)
signature.asc
Description: Digital signature
Bug#564258: xserver-xorg-input-evdev: evdev on accelerometers is broken
Hello Julien, Julien Cristau [2010-01-19 22:43 +0100]: It looks like this because I changed it after people started reporting this problem, and ubuntu synced it :) See http://git.debian.org/?p=pkg-xorg/driver/xserver-xorg-input-evdev.git;a=commitdiff;h=d293b7ae066ab55540809b99ba770174dfca3bc7 Ah, splendid. Seems you're way ahead of me :) ? If so, my recommendation would be to drop this. We don't need X.org to listen to obscure devices like your accelerometer or headphone jacks. Button-type devices like lids are already covered by ID_INPUT_KEY (which is not the same as _KEYBOARD, NB). Hrm. This means with inputclasses we'll need one inputclass section for each device type. Or tag the devices we don't want to handle somehow in the udev db. I'm afraid I don't quite understand this. AFAIUI, the input classes will look directly at the udev properties, and the x11_* properties will be dropped completely. But Peter's proposal [1] already has MatchIsKeyboard, MatchIsPointer, MatchIsJoystick, and the like. So am I missing something, and this isn't enough for some reason? (For this particular bug we should probably not do the evdev-catchall thing, though.) Thanks, Martin [1] http://who-t.blogspot.com/2010/01/new-configuration-world-order.html -- Martin Pitt| http://www.piware.de Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org) -- To UNSUBSCRIBE, email to debian-x-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: Bug#300185: FTBFS in experimental
Hi! Andreas Barth [2005-03-18 10:29 +0100]: Automatic build of postgresql-7.4_1:7.4.7-1 on notos by sbuild/powerpc 1.170.5 /usr/include/tcl8.4/tk.h:96:29: X11/Xlib.h: No such file or directory I really don't understand this. This seems to happen at random, affects the Sarge/Sid version as well and seems to affect only some architectures. Xlib.h is in libx11-dev, tk8.4-dev needs it and depends on it. The build log shows that the package is installed: Selecting previously deselected package libx11-dev. Unpacking libx11-dev (from .../libx11-dev_4.3.0.dfsg.1-12_powerpc.deb) ... [...] Setting up libx11-dev (4.3.0.dfsg.1-12) ... [...] Previous psotgresql versions have built fine with exactly the same build dependencies. Maybe tk8.4-dev or XFree86 introduced a change which could explain this weirdness? Thanks for any idea, Martin -- Martin Pitt http://www.piware.de Ubuntu Developerhttp://www.ubuntulinux.org Debian GNU/Linux Developer http://www.debian.org signature.asc Description: Digital signature
Bug#149631: [ati/atimisc] DoS attack possible when viewing absurdly huge fonts on Mach64 GT rev 65
Hi! On 2002-10-30 17:04 +0100, Juliusz Chroboczek wrote: This issue should hopefully be fixed upstream in 4.3.0. Is this really the case? If so, this scary bug should be closed. There will be no fix for 4.2. No problem, this version is history. Thanks and have a nice day! Martin -- Martin Pitt Debian GNU/Linux Developer [EMAIL PROTECTED] [EMAIL PROTECTED] http://www.piware.de http://www.debian.org signature.asc Description: Digital signature
Bug#240351: xserver-xfree86: Alt-Gr does not work anymore
Hi Christian! On 2004-03-27 22:24 +0100, Christian Guggenberger wrote: Do you use login managers like xdm, kdm or gdm? if so, let us know, if things like []}\ work there, or not? Oh, I forgot to mention: I use xdm and I also tried that out there: it does not work, neither in xdm, nor in xterm, eterm, mozilla, or any other program I've tried. At least it is consistent :-) Thanks and have a nice day! Martin, who needs to use mouse cutpaste to get some characters in X -- Martin Pitt Debian GNU/Linux Developer [EMAIL PROTECTED] [EMAIL PROTECTED] http://www.piware.de http://www.debian.org signature.asc Description: Digital signature
Bug#240351: xserver-xfree86: Alt-Gr does not work anymore
Hi again! After purging and reinstalling several X11-related packages, I finally found that reinstalling xbase-clients solves the problem. So this bug seems to be a transition problem, therefore I don't close the bug. Unfortunately I did not save the files of the old package, so I guess debugging will be quite hard. So if you don't have a clue what the cause could have been (maybe an old compiled keyboard map?) and I'm the only one who had this problem, then feel free to close this bug. At least I'm happy that I can enter all characters again :-) (Well, all but the Euro sign) Thanks and have a nice weekend! Martin -- Martin Pitt Debian GNU/Linux Developer [EMAIL PROTECTED] [EMAIL PROTECTED] http://www.piware.de http://www.debian.org signature.asc Description: Digital signature
Bug#240351: xserver-xfree86: Alt-Gr does not work anymore
Hi! On 2004-03-28 0:13 +0100, Christian Guggenberger wrote: Have you ever had the prerelease packages from experimental installed ( or DanielS' unofficial 4.3 packages)? That could explain transition problems. About two months ago I had a Radeon 9200 Card to try out and I installed 4.3.0 from experimental. But since it did not work, I returned the card, purged the experimental packages and reinstalled 4.2.x (the Sid version of that time) again. It may very well be that the purge left some files over that still remained on my system. If the experimental package was the reason then I think we can forget about this bug. Thanks for your help and have a nice day! Martin -- Martin Pitt Debian GNU/Linux Developer [EMAIL PROTECTED] [EMAIL PROTECTED] http://www.piware.de http://www.debian.org signature.asc Description: Digital signature
Bug#240351: xserver-xfree86: Alt-Gr does not work anymore
-xfree86/multiple_possible_x-drivers: * xserver-xfree86/config/inputdevice/keyboard/model: pc105 xserver-xfree86/config/write_dri_section: true xserver-xfree86/config/device/driver: nv xserver-xfree86/config/monitor/vert-refresh: 43-75 * xserver-xfree86/config/display/default_depth: 24 xserver-xfree86/config/inputdevice/mouse/zaxismapping: true * xserver-xfree86/config/display/modes: 1280x1024, 1280x960, 1024x768, 800x600, 640x480 xserver-xfree86/config/device/bus_id_error: xserver-xfree86/config/modules: GLcore, bitmap, dbe, ddc, dri, extmod, freetype, glx, int10, record, speedo, type1, vbe * xserver-xfree86/config/inputdevice/keyboard/options: xserver-xfree86/config/nonnumeric_string_error: xserver-xfree86/config/inputdevice/mouse/protocol: ImPS/2 shared/multiple_possible_x-servers: xserver-xfree86/config/null_string_error: xserver-xfree86/config/monitor/range_input_error: * xserver-xfree86/autodetect_video_card: true * xserver-xfree86/config/inputdevice/keyboard/variant: nodeadkeys * xserver-xfree86/config/inputdevice/mouse/port: /dev/input/mice xserver-xfree86/config/write_files_section: true * xserver-xfree86/autodetect_monitor: true -- Martin Pitt Debian GNU/Linux Developer [EMAIL PROTECTED] [EMAIL PROTECTED] http://www.piware.de http://www.debian.org signature.asc Description: Digital signature
